Re: Private company and code salvation

[Jaroslav Hajek]

On Mon, Sep 29, 2008 at 12:31 PM, David Bateman
<address@hidden> wrote:
> Jaroslav Hajek wrote:
>>
>> David, forgive me if I'm being too naive, but if any company wants to
>> use Octave "commercially" and wishes an OO LGPL API to exist, they're
>> obviously free and welcome to donate money specifically for such a
>> project. I'm sure the community would answer such a demand.
>> It seems to me, though, that the community itself, developing GPL
>> software,  would not benefit from such an interface, which obviously
>> raises a question why should they (we) bother. Of course, a
>> goal-specific donation *is* a good reason.
>>
>
> All of my thoughts come from the question of why Octave isn't seeing a
> significant amount of funding given its large user base. And from that what
> can and should be done to get funding for Octave and in particular what
> would need to be done to finance a company formed to supply paid support for
> Octave.
>
> To me the question is not one of whether a company would finance such an
> API. Given that the Octave community would accept such an API, which is
> still not certain,

maybe, but I don't see why we should object...

> and someone is willing to go and sell the use of Octave
> to commercial partners,

agreed, Octave is little known to potential commercial users.

>then yes I fully believe that money could be found
> to write it. Companies don't just throw their money at open source projects.

No. They throw it if there's something they want to be done (like
creating a LGPL API so that they can use it).
On the contrary, if the software already fills their needs, there's
not much reason for them to donate.

> Someone has to go and ask for it, and give good reasons why giving that
> money to the open source project is good for the company doing it.

Yes, precisely. In fact, once a company realises that they would
benefit from using Octave (say, instead of Matlab), and that the only
obstacle is the missing LGPL API, it will probably be clear to them
that investing money to get that last obstacle removed is a good idea.
On the other hand, if the LGPL API is already there, they will just
use it.

> The thought behind the need for such an API is that any support company
> formed to supply support to users paying for it, and incidentally also
> contributing code to Octave, would need to fight the argument that any code
> a commercial partner wrote using Octave would have to be GPLed, which would
> significantly limit where funding might be found. We as a community are
> facing the issue at the moment of how to finance the development of Octave,
> and we need to ask ourselves why the money isn't already rolling in, and
> question our own role in the fact that it isn't..

I think we should simply get Octave more known.


>> In commercial software, you pay for using existing functionality,
>> while in open source software, you typically pay (donate) for
>> functionality that you wish to be implemented (and then you can use it
>> freely).
>>
>
> That still doesn't mean that someone doesn't have to go and ask for that
> "donation" and in the process "sell" the company on the use of Octave.

Yeah, OK. But my understanding was that you wanted the functionality
to be done first and *then* you come to the company and ask for the
donation as a reward. Well, it's possible, but not a good idea IMHO.

In other words, I think it's harder for anyone to justify (to a CEO,
for example) a donation for Octave by telling "Octave's great, and I'm
using it a lot, and for free, so why don't we donate some money to
those guys?".
If instead you say "Octave's great, and I'm using it a lot, but I
would do much better with this particular feature that isn't there, so
why don't we donate some money so that it gets implemented?",
then IMHO you have a far better chance to succeed and actually get the
money. And the LGPL ABI is an excellent candidate for such a feature.
People can use Octave for commercial purposes, but they cannot deploy
DLD functions without source. The GPL developers don't miss this
option. The commercial developers do, so it's a good argument to
convince their bosses that donating some money for it is a good idea.


> There are other reasons why perhaps creating a paid support company for
> Octave isn't necessarily a great idea. The major one being that any such
> entity is a logical target for a patent lawsuit on the part of Mathworks.
> Mathworks aren't hesitant in using patent law to suppress competition (cf
> .http://dockets.justia.com/docket/court-txedce/case_no-6:2006cv00334/case_id-97908/)
> But forming a paid support company has to be considered as one option of
> getting funding for the development of Octave in the current situation.
>

I don't understand the laws much, but IMHO a company based in Europe
would be in less danger as most Mathworks patents seem to be US only
and in general laws in EU are more strict about software patents.

> Cheers
> D.
>
> --
> David Bateman                                address@hidden
> Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) Parc Les
> Algorithmes, Commune de St Aubin    +33 6 72 01 06 33 (Mob) 91193
> Gif-Sur-Yvette FRANCE                  +33 1 69 35 77 01 (Fax)
> The information contained in this communication has been classified as:
> [x] General Business Information [ ] Motorola Internal Use Only [ ] Motorola
> Confidential Proprietary
>
>



-- 
RNDr. Jaroslav Hajek
computing expert
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz